A method for linearizing calibration of a continuous casting mold liquid level control receiver

Abstract

The application discloses a method for linearizing and calibrating a liquid level control receiver of a continuous casting crystallizer, and belongs to the technical field of automatic control of continuous steel casting. The method draws a demarcation point in a calibration interval from a copper pipe mouth of the crystallizer, point by point obtains a radiation measurement value of the receiver to form an original calibration data sequence, and records a maximum value as N0; calculates a radiation difference value between adjacent demarcation points to determine a resolution degree distribution of each interval; takes a preset proportion of N0 as a tolerance value a, takes a preset multiple of an average difference value obtained by equally dividing N0 according to the number of intervals as a maximum tolerance difference value b, accumulates the radiation difference value from an initial calibration value lower than a, and makes the difference value monotonously increase from a full shielding end to a normal liquid level position to a maximum value not more than b, and then monotonously decrease to a non-shielding end, so that an S-shaped calibration curve is obtained; the S-shaped calibration curve is multiplied by a correction coefficient to make the maximum value not more than N0, and then input into a secondary instrument, so that the liquid level control of the crystallizer is realized. According to the application, the liquid level fluctuation is reduced from ±3mm to ±2mm.

Description

Technical Field

[0001] This invention relates to the field of automatic control technology for continuous steel casting, and in particular to a method for linearizing and calibrating the receiver for liquid level control in a continuous casting crystallizer. Background Technology

[0002] The molten steel level control in continuous casting machine molds generally adopts a radiation source-type molten steel level control system. Its molten steel detection section consists of secondary instruments, a receiver, a radiation source, and high-temperature resistant connecting components. The radiation source is located on one side of the mold, and the receiver is located on the other side. The radiation emitted by the source passes through the mold and is received by the receiver. The higher the molten steel level in the mold, the more radiation is blocked, and the smaller the radiation measurement value received by the receiver. The secondary instruments calculate the molten steel level position in the mold based on the radiation measurement value output by the receiver, and control the opening of the tundish stopper or sliding gate accordingly, achieving closed-loop control of the mold level.

[0003] To ensure the secondary instrument can accurately calculate the liquid level from the radiation measurement values, the receiver of the continuous casting machine needs to be linearized and calibrated before billet production. The existing linearization calibration method involves simulating the molten steel level with a calibration object inside the crystallizer. Starting 5 cm from the copper tube opening of the crystallizer, calibration positions are set every 1 cm until 15 cm from the opening, forming 11 dividing points and 10 dividing intervals. The radiation measurement value of the receiver at each position is recorded as the N value for that point, and written into the secondary instrument in the order of N0 to N10, thus establishing the correspondence between the radiation measurement value and the molten steel level. At N0, the calibration object is unobstructed, resulting in the highest radiation measurement value; at N10, the calibration object is completely obstructed, resulting in the lowest radiation measurement value.

[0004] The existing methods described above merely passively record the actual radiation measurements of the equipment without intervening in their distribution across the various intervals. Radiation sources and receivers operate in high-temperature, high-humidity, and dusty environments for extended periods. Source intensity attenuation and receiver performance variations can cause uneven distribution of radiation counts across the intervals. For example, with a total radiation value interval of 20,000, ideally it should be evenly divided into intervals of 2,000, but actual calibration results often deviate significantly from this distribution.

[0005] The difference in radiation measurements between two adjacent dividing points, i.e., Nn+1-Nn, determines the resolution of that dividing interval: the larger the radiation difference, the greater the change in radiation measurement caused by the same magnitude of liquid level change, the stronger the secondary instrument's ability to resolve liquid level changes within that interval, and the stronger its ability to resist field signal disturbances; conversely, the smaller the radiation difference, the lower the resolution of that interval, and even small signal disturbances may be misjudged as liquid level changes by the secondary instrument.

[0006] Actual measurements show that the resolution distribution obtained by existing calibration methods is exactly the opposite of production requirements: in the 60% to 80% range where the molten steel level frequently remains, the radiation difference between adjacent dividing points is only 3000 to 3500; while in the 10% to 40% range, which is rarely used in production, the radiation difference between adjacent dividing points reaches 5000 to 6000. This results in the lowest resolution and weakest resistance to disturbances when the molten steel is near its level position. The level signal output by the secondary instrument fluctuates greatly, causing the stopper rod or sliding nozzle to move frequently. The liquid level fluctuation in the crystallizer reaches ±3mm, which can easily form creases on the surface of the billet and induce surface defects such as cracks and corner cracks. In severe cases, it can lead to the billet being rejected. Summary of the Invention

[0007] The technical problem to be solved by the present invention is to overcome the shortcomings of the prior art and provide a method for linearizing and calibrating the liquid level control receiver of a continuous casting crystallizer.

[0008] To solve the above technical problems, the technical solution of the present invention is as follows: This invention provides a method for linearizing and calibrating a liquid level control receiver in a continuous casting crystallizer, comprising the following steps: S1. Within the preset calibration range from the copper tube opening of the crystallizer, divider points are marked, and the radiation measurement values ​​of the receiver are obtained point by point to obtain the original calibration data sequence; among them, the maximum measurement value is the measurement value N0 at the unobstructed end when the calibration object is unobstructed, and the minimum measurement value is the measurement value N10 at the fully obstructed end when the calibration object is fully obstructed. S2. Calculate the radiation difference between adjacent dividing points from the original calibration data sequence to obtain the original difference sequence and determine the resolution distribution of each dividing interval; S3. Based on the resolution distribution, take the preset ratio of N0 as the tolerance value a for the fully shielded end, and take the preset multiple of the average difference obtained by dividing N0 into equal intervals as the maximum tolerance difference b. Determine the initial calibration value of the fully shielded end below a, and accumulate the radiation difference point by point from it, so that the radiation difference increases monotonically from the fully shielded end to the position where the molten steel is normally liquid and reaches a maximum value not exceeding b, and decreases monotonically from that position to the unshielded end, thus obtaining the adjusted sequence that constitutes the S-shaped calibration curve. S4. Multiply the adjusted sequence by the correction factor so that the maximum value of the multiplication does not exceed N0, and obtain the final calibration value and input it into the secondary instrument.

[0009] The beneficial effects of this invention are: (1) This invention determines the original resolution distribution of each segmented interval, and then makes the radiation difference monotonically increase from the fully shielded end to the molten steel constant liquid level position and reach its maximum value, and then monotonically decrease from that position to the unshielded end, thereby concentrating the limited radiation measurement range of the receiver towards the molten steel constant liquid level position. Since the radiation difference within the segmented interval directly determines the resolution of that interval, a liquid level change of the same amplitude near the molten steel constant liquid level position will cause a larger change in radiation measurement value. The secondary instrument can then distinguish the actual liquid level change from the field signal disturbance, avoiding misjudging minor disturbances as liquid level fluctuations. After calibration using this invention, the fluctuation amplitude of the crystallizer liquid level within the effective control range of about 70% decreased from ±3mm to ±2mm.

[0010] (2) This invention introduces two constraints: a tolerance value for the fully obscured end and b maximum tolerance difference. These two constraints, one at the bottom and one at the top, jointly define the feasible domain of reconstruction. a is determined by a preset proportion of the measurement value N0 at the unobscured end, requiring the initial calibration value at the fully obscured end to be lower than this value. This ensures sufficient margin for radiation measurement values ​​at the lower end of the sequence, which can be used when increasing the radiation difference in the middle interval. b is determined by a preset multiple of the average difference obtained by dividing N0 equally among the intervals, requiring the radiation difference in any interval to not exceed this upper limit. This avoids excessive expansion of the difference at the point where molten steel is constantly at its liquid level, which could lead to a collapse in the resolution of other intervals or local distortion of the curve. These two constraints ensure that the reconstructed sequence remains continuous, monotonous, and smooth throughout the entire range.

[0011] (3) The maximum value of the reconstructed data sequence in this invention may exceed the maximum radiation measurement value N0 that the receiver can actually measure. If it is directly written into the secondary instrument, there will be calibration values ​​that the receiver cannot output at the high end of the sequence. When the molten steel level is low, the secondary instrument cannot obtain the corresponding radiation measurement value, and the calibration relationship will fail in this segment. Therefore, the entire reconstructed sequence is multiplied by a correction factor so that the maximum value after multiplication does not exceed N0. Since all calibration values ​​are multiplied by the same factor, the relative proportion of the radiation difference between the intervals within the sequence remains completely unchanged, the established resolution distribution pattern is preserved as is, and the entire curve is compressed back into the actual range of the receiver, which can be directly implemented on existing secondary instruments.

[0012] (4) All operations of this invention are completed in the linearization calibration process before the continuous casting machine produces the billet. Only the calibration value written to the secondary instrument is changed. It does not require replacement of the radiation source, receiver or secondary instrument, addition of detection elements or modification of high-temperature resistant connection components. Therefore, it does not increase equipment investment or on-site construction. When the equipment continues to age and the resolution distribution shifts again, the calibration can be re-executed according to the same process, which has good repeatability. At the same time, since the fluctuation range of the liquid level in the crystallizer is reduced from ±3mm to ±2mm, the creases on the surface of the billet caused by the violent fluctuation of the liquid level are significantly reduced. The defects such as cracks and corner cracks induced by the creases are reduced accordingly, the number of billet scraps decreases, and the yield is improved. Attached Figure Description

[0013] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0014] Figure 1 A schematic flowchart illustrating the method for linearizing and calibrating the liquid level control receiver in a continuous casting crystallizer provided by the present invention; Figure 2 This is a schematic diagram of the original measured values ​​in the embodiment; Figure 3 This is a line chart showing the difference in the original data in the example; Figure 4 This is a schematic diagram of the optimized difference curve in the example. Detailed Implementation

[0015] To make the content of this invention easier to understand, the invention will be further described in detail below with reference to specific embodiments and accompanying drawings.

[0016] This embodiment uses a No. 6 continuous casting machine in a steel plant as an example. This continuous casting machine is an R10 arc-shaped continuous casting machine, mainly producing 165mm square and 220mm x 165mm small square billets. Its crystallizer liquid level control adopts a radiation source-type molten steel liquid level control system. The molten steel detection section consists of secondary instruments, a receiver, a radiation source, and high-temperature resistant connecting components. In this embodiment, the molten steel is typically positioned at 70% of the crystallizer liquid level.

[0017] See Figure 1 This application provides a method for linearization calibration of a liquid level control receiver in a continuous casting crystallizer, which specifically includes the following steps: Step S1: Define the dividing points within the preset calibration range from the copper tube opening of the crystallizer, and obtain the radiation measurement values ​​of the receiver point by point to obtain the original calibration data sequence.

[0018] Specifically, this step is performed before the continuous casting machine produces the billet, with the aim of obtaining the actual radiation measurement distribution of the receiver under the current equipment condition, as a benchmark for subsequent reconstruction.

[0019] Inside the crystallizer, a calibration object is used to simulate the surface of molten steel, and dividing points are marked within a preset calibration range from the copper tube opening of the crystallizer. In this embodiment, the preset calibration range is the interval between 5 cm and 15 cm from the copper tube opening of the crystallizer; dividing points are set every 1 cm starting from 5 cm from the copper tube opening of the crystallizer, forming a total of 11 dividing points arranged in the order of N0 to N10. Two adjacent dividing points constitute a dividing interval, for a total of 10 dividing intervals.

[0020] The calibration object is placed sequentially at each dividing point, and the radiation measurement values ​​output by the receiver are read point by point, recorded in the order of N0 to N10, to obtain the original calibration data sequence. The end where the calibration object is unobstructed receives the most radiation and has the highest radiation measurement value is the unobstructed end, and its measurement value is recorded as N0. The end where the calibration object is completely obstructed receives the least radiation and has the lowest radiation measurement value is the fully obstructed end, which is N10 in this embodiment. The unobstructed end corresponds to the 0% position of the crystallizer liquid level, and the fully obstructed end corresponds to the 100% position of the crystallizer liquid level.

[0021] The original calibration data sequence obtained in this embodiment is as follows: Figure 2 As shown.

[0022] Step S2: Calculate the radiation difference between adjacent dividing points from the original calibration data sequence to obtain the original difference sequence and determine the resolution distribution of each dividing interval.

[0023] Specifically, the purpose of this step is to convert the absolute measurement value obtained in step S1 into a relative quantity that can reflect the resolution of each interval, so as to determine how the radiation count is allocated across the entire range, which segment is compressed, and which segment is wasted.

[0024] The radiation difference Nn+1-Nn between adjacent separators is calculated one by one from the original calibration data sequence, resulting in 10 radiation differences, forming the original difference sequence. The original difference sequence is then plotted as a line graph according to the separator intervals, as shown below. Figure 3 As shown.

[0025] The larger the radiation difference within a certain interval, the more radiation counts the secondary instrument can use to distinguish liquid level changes within that interval, resulting in higher resolution and stronger resistance to field signal disturbances; conversely, the smaller the difference, the lower the resolution. Therefore, the resolution distribution of each interval can be directly read from the difference line graph.

[0026] Depend on Figure 3As can be seen, in this embodiment, the radiation difference in the 60% to 80% range where the molten steel level frequently remains (segments 7 and 8) is only 3000 and 3500, respectively, which is the lowest level in the entire sequence. However, in the 10% to 40% range (segments 2 to 4), which is rarely used in production, the radiation difference reaches 6000, 5000, and 5000. The resolution distribution is exactly the opposite of the production requirements, which is the root cause of the large fluctuations in the molten steel level at the current position, and also the basis for the reconstruction in step S3.

[0027] Step S3: Based on the resolution distribution, take the preset ratio of N0 as the tolerance value a for the fully shielded end, and take the preset multiple of the average difference obtained by dividing N0 into equal intervals as the maximum tolerance difference b. Determine the initial calibration value of the fully shielded end that is lower than a. Starting from this value, accumulate the radiation difference point by point, so that the radiation difference increases monotonically from the fully shielded end to the position where the molten steel is normally at the liquid level and reaches a maximum value not exceeding b. Then, from this position to the unshielded end, it decreases monotonically, thus obtaining the adjusted sequence that constitutes the S-shaped calibration curve.

[0028] This step is the core of the invention. The idea is to reclaim radiation counts from less important production intervals, while ensuring the total radiation measurement range cannot be increased, and then allocate these counts to the intervals where the molten steel is typically at its liquid level. To ensure the reconstruction result is feasible and smooth, two constraints are first set, and then a new sequence is generated by accumulating points from the fully obscured end.

[0029] This step specifically includes the following sub-steps: Step S3-1: Determine two constraints. A preset proportion of the unobstructed end measurement value N0 is used as the tolerance value 'a' for the fully obstructed end. The reconstructed calibration value of the fully obstructed end must not exceed 'a'. In this embodiment, the preset proportion is 1 / 15, i.e., a = N0 / 15 = 48000 / 15 = 3200. A preset multiple of the average difference obtained by dividing N0 into equal intervals is used as the maximum tolerance difference value 'b'. The radiation difference of any interval after reconstruction must not exceed 'b'. In this embodiment, the number of intervals is 10, and the preset multiple is 1.5, i.e., b = 1.5 * (N0 / 10) = 1.5 * 4800 = 7200.

[0030] The function of 'a' is to force the lower end of the sequence to make the calibration value of the fully occluded end lower than the radiation measurement value of the corresponding fully occluded end in the original calibration data sequence (originally 6700 in this embodiment), thereby freeing up the radiation measurement value margin at the lower end for use when increasing the radiation difference in the middle interval; the function of 'b' is to limit the upper limit of radiation counts that a single segment can occupy, to avoid excessive expansion of the radiation difference at the liquid level position of molten steel, which would squeeze the resolution of other intervals or cause local distortion of the curve.

[0031] Step S3-2: Determine the initial calibration value of the fully shielded end. The initial calibration value of the fully shielded end is determined within a range lower than 'a', serving as the starting point for point-by-point accumulation. In this embodiment, 2600 is used, which is lower than 'a' (3200) and also lower than the radiation measurement value of the fully shielded end in the original calibration data sequence (6700).

[0032] Step S3-3: Accumulate the radiation difference value point by point starting from the fully shielded end. Starting from the initial calibration value at the fully shielded end, accumulate the radiation difference value point by point along the direction from the fully shielded end to the unshielded end to obtain the calibration values ​​of each step in the adjusted sequence. The radiation difference value applied during the accumulation process must satisfy the following: monotonically increase from the fully shielded end to the position where the molten steel is normally at its liquid level, and obtain the maximum value in the entire sequence at that position, and this maximum value does not exceed b; monotonically decrease from that position to the unshielded end.

[0033] It should be noted that the above monotonically increasing means linearly increasing and monotonically decreasing means linearly decreasing. That is, the quadratic difference between adjacent radiation differences remains constant, which is reflected in the difference line graph as the line connecting two adjacent points remains a straight line. Furthermore, the radiation difference at the unshielded end is equal to the radiation difference at the fully shielded end, meaning that the two endpoints of the difference line are on the same horizontal line.

[0034] In this embodiment, the molten steel is typically positioned at 70% of the crystallizer's liquid level, corresponding to dividing point N7. Starting from N10 (2600), values ​​are sequentially increased by 3400, 4700, and 6100 to obtain N9, N8, and N7. The radiation difference reaches its maximum value of 7200 in the dividing interval between N6 and N7, which is exactly less than b (7200). Subsequently, values ​​are sequentially increased by 7000, 6500, 6000, 5500, 5000, and 4400, with the radiation difference decreasing linearly to the unobstructed end. The resulting adjusted sequence and its radiation difference and quadratic difference are as follows: Figure 4 As shown.

[0035] Depend on Figure 4 As can be seen, the radiation difference in the adjusted sequence reaches 7200 in the interval where the molten steel is at its normal liquid level, more than double the 3000 of the original difference sequence at that location. Meanwhile, the radiation difference in the 10% to 40% interval decreases from 6000, 5000, 5000 to around 5000, 5500, 6000 and continues to decrease towards the unobstructed end, indicating that the recovered radiation counts have been transferred to the normal liquid level position of the molten steel. The two endpoints of the difference line are 4400 and 3400, located on roughly the same horizontal line.

[0036] The adjusted sequence forms an S-shaped calibration curve: it changes gently on both sides of the unobstructed end and the fully obstructed end, and changes most steeply near the position where the molten steel is usually at the liquid level, that is, the radiation measurement value corresponding to the unit liquid level change near this position changes the most.

[0037] Step S3-4: Verification. After accumulation, verify whether the adjusted sequence meets two constraints: the calibration value of 2600 at the fully occluded end is lower than a (3200); the maximum value of the radiation difference between each interval, 7200, does not exceed b (7200). If both are satisfied, proceed to step S4.

[0038] Step S4: Multiply the adjusted sequence by the correction factor so that the maximum value of the multiplication does not exceed N0, obtain the final calibration value and input it into the secondary instrument.

[0039] The maximum value of the adjusted sequence obtained in step S3 is 58400, which exceeds the maximum radiation measurement value N0 (48000) that the receiver can actually measure. If it is directly written into the secondary instrument, there will be a calibration value at the high end of the sequence that the receiver cannot output at all. When the molten steel level is low, the secondary instrument cannot obtain the corresponding radiation measurement value, and this calibration relationship will be invalid.

[0040] Therefore, the adjusted sequences are uniformly multiplied by a correction factor so that the maximum value of the multiplication does not exceed N0. In this embodiment, the correction factor is 0.8, and the maximum value after multiplication is 58400 × 0.8 = 46720, which does not exceed 48000. This approximates the original measured N0 value of 48000 in a way that does not exceed the N0 value. After multiplying each sequence from N0 to N10 by the same factor of 0.8, the resulting new sequences from N0 to N10 are the adjusted calibration values.

[0041] Since all calibration values ​​are multiplied by the same correction factor, the relative proportion of the radiation difference between the intervals within the sequence remains unchanged. The resolution allocation pattern established in step S3 is completely preserved, and the entire S-shaped calibration curve is compressed proportionally only in the vertical direction to the actual range of the receiver.

[0042] Input the final calibration values ​​into the secondary instrument in the order of N0 to N10 to complete this linearization calibration.

[0043] After this embodiment was put into use, the fluctuation range of the liquid level in the crystallizer was within an effective control range of about 70%, which decreased from ±3mm before calibration to ±2mm; the creases on the surface of the billet caused by the fluctuation of the liquid level were significantly reduced, and the defects such as cracks and corner cracks induced by the creases were reduced accordingly, resulting in a decrease in the number of billets rejected.

[0044] In addition to the above embodiments, the present invention may have other implementation methods; all technical solutions formed by equivalent substitution or equivalent transformation fall within the protection scope claimed by the present invention.

Claims

1. A method for linearizing and calibrating a liquid level control receiver in a continuous casting crystallizer, characterized in that, Includes the following steps: S1. Within the preset calibration range from the copper tube opening of the crystallizer, divider points are marked, and the radiation measurement values ​​of the receiver are obtained point by point to obtain the original calibration data sequence; among them, the maximum measurement value is the measurement value N0 at the unobstructed end when the calibration object is unobstructed, and the minimum measurement value is the measurement value N10 at the fully obstructed end when the calibration object is fully obstructed. S2. Calculate the radiation difference between adjacent dividing points from the original calibration data sequence to obtain the original difference sequence and determine the resolution distribution of each dividing interval; S3. Based on the resolution distribution, take the preset ratio of N0 as the tolerance value a for the fully shielded end, and take the preset multiple of the average difference obtained by dividing N0 into equal intervals as the maximum tolerance difference b. Determine the initial calibration value of the fully shielded end that is lower than a. Starting from this value, accumulate the radiation difference point by point, so that the radiation difference increases monotonically from the fully shielded end to the position where the molten steel is normally at the liquid level and reaches a maximum value not exceeding b. Then, it decreases monotonically from this position to the unshielded end, thus obtaining the adjusted sequence that constitutes the S-shaped calibration curve. S4. Multiply the adjusted sequence by the correction coefficient so that the maximum value of the multiplication does not exceed N0, and obtain the final calibration value and input it into the secondary instrument.

2. The method for linearizing and calibrating the liquid level control receiver of a continuous casting crystallizer according to claim 1, characterized in that, In step S1, the preset calibration interval is the interval between 5cm from the copper tube opening of the crystallizer and 15cm from the copper tube opening of the crystallizer. The dividing point is set every 1cm starting from 5cm from the copper tube opening of the crystallizer, forming a total of 11 dividing points and 10 dividing intervals arranged in the order of N0 to N10, wherein N10 is the fully shielded end.

3. The method for linearizing and calibrating the liquid level control receiver of a continuous casting crystallizer according to claim 1, characterized in that, In step S3, the monotonically increasing radiation difference is linearly increasing, the monotonically decreasing is linearly decreasing, and the quadratic difference between adjacent radiation differences remains constant during the point-by-point accumulation process.

4. The method for linearizing and calibrating the liquid level control receiver of a continuous casting crystallizer according to claim 2, characterized in that, In step S3, the molten steel is usually positioned at 70% of the liquid level in the crystallizer, and the 70% liquid level in the crystallizer corresponds to N7 in the dividing point.

5. The method for linearizing and calibrating the liquid level control receiver of a continuous casting crystallizer according to claim 2, characterized in that, In step S3, the tolerance value of the fully occluded end is a = N0 / 15, and the maximum tolerance difference is b = 1.5 × (N0 / 10).

6. The method for linearizing and calibrating the liquid level control receiver of a continuous casting crystallizer according to claim 1, characterized in that, The calibration value corresponding to the fully shielded end in the adjusted sequence is lower than the radiation measurement value corresponding to the fully shielded end in the original calibration data sequence.

7. The method for linearizing and calibrating the liquid level control receiver of a continuous casting crystallizer according to claim 5, characterized in that, The value of N0 is 48000, the tolerance value a of the full occlusion end is 3200, the maximum tolerance difference value b is 7200, and the initial calibration value of the full occlusion end is 2600.

8. The method for linearizing and calibrating the liquid level control receiver of a continuous casting crystallizer according to claim 7, characterized in that, The maximum value of the adjusted sequence obtained in step S3 is 58400, the correction coefficient in step S4 is 0.8, and the maximum value after multiplication is 46720.

9. The method for linearizing and calibrating the liquid level control receiver of a continuous casting crystallizer according to claim 3, characterized in that, In the adjusted sequence, the radiation difference corresponding to the unobstructed end is equal to the radiation difference corresponding to the fully obstructed end.

Images